PREDICTIVE ANALYTICS BASED ON ARTIFICIAL INTELLIGENCE AS A TOOL FOR MINIMIZING CUSTOMER CHURN IN DIGITAL SERVICES
Abstract
Customer retention has become a critical indicator of financial stability for digital services amid economic digitalization and subscription model proliferation. Growing competition and market saturation lead to increased churn rates, posing systemic threats to business viability. Traditional prediction methods prove insufficient for large volumes of non-functional data. The purpose of this article is to substantiate the role of predictive analytics based on artificial intelligence as a tool for minimizing customer churn and to develop a conceptual model for its implementation in digital services. The study employs methods of systematization, comparative analysis of machine learning algorithms, and conceptual modeling. Recent scientific publications from 2022 to 2025 were analyzed to identify current trends. A 5-layer conceptual model was developed, comprising Data Layer, Analytics Layer, Decision Layer, Action Layer, and Risk Management Layer with feedback loop for adaptive retraining. The analysis established that for Ukrainian digital services, the optimal balance lies between accuracy and interpretability. Ensemble methods demonstrate the best ratio for structured data, while hybrid deep learning architectures suit large platforms. Simpler methods achieve accuracy exceeding 85% given quality data preparation. Three groups of implementation risks were identified: ethical risks including data confidentiality and algorithmic bias, managerial risks including staff resistance and personnel deficit, and technical risks including data quality and conceptual drift. The practical value lies in the developed model and comparative algorithm characteristics for informed selection of AI solutions. This enables transition from reactive to proactive loyalty management and increases customer lifetime value. The proposed model creates a methodological foundation for further research and practical implementation of AI in Ukrainian digital services. Future research should focus on empirical validation using local datasets and integration of explainable AI tools.
References
Geiler L., Affeldt S., Nadif M. A survey on machine learning methods for churn prediction. International Journal of Data Science and Analytics. 2022. Vol. 14, № 3. P. 217–242. DOI: https://doi.org/10.1007/s41060-022-00312-5
Imani M., Joudaki M., Beikmohammadi A., Arabnia H. R. Customer Churn Prediction: A Systematic Review of Recent Advances, Trends, and Challenges in Machine Learning and Deep Learning. Machine Learning and Knowledge Extraction. 2025. Vol. 7, № 3. Art. 105. URL: https://www.mdpi.com/2504-4990/7/3/105 (дата звернення: 07.04.2026). DOI: https://doi.org/10.3390/make7030105
Khattak A., Mehak Z., Ahmad H., Asghar M. U., Asghar M. Z., Khan A. Customer churn prediction using composite deep learning technique. Scientific Reports. 2023. Vol. 13, Art. 17294. URL: https://www.nature.com/articles/s41598-023-44396-w#citeas (дата звернення: 07.04.2026). DOI: https://doi.org/10.1038/s41598-023-44396-w
Papa A., Shemet Y., Yarovyi A., Vahovska L. Development of information technology for analyzing the customer churn of a telecommunication company. Information and Control Systems. 2022. Vol. 2, № 2(64). P. 11–15. DOI: https://doi.org/10.15587/2706-5448.2022.255861
Reichheld F. F., Sasser W. E. Jr. Zero defections: Quality comes to services. Harvard Business Review. 1990. Vol. 68, № 5. P. 105–111.
Євтушенко Н. М., Кривенко Ю. В., Стеценко Д. І. Цифрові технології у страхуванні. Міжнародний науковий журнал «Грааль науки». 2024. № 43. С. 105–114. DOI: https://doi.org/10.36074/grail-of-science.06.09.2024.011
Суружіу І. С., Леонова С. В., Таранський І. П. Штучний інтелект у процесі трансформації інтегрованих маркетингових комунікацій інноваційно активних підприємств. Маркетинг і цифрові технології. 2024. Т. 8, № 4. С. 45–58. DOI: https://doi.org/10.5281/zenodo.14032942
Geiler L., Affeldt S., Nadif M. (2022) A survey on machine learning methods for churn prediction. International Journal of Data Science and Analytics, vol. 14, no 3, pp. 217–242. DOI: https://doi.org/10.1007/s41060-022-00312-5
Imani M., Joudaki M., Beikmohammadi A., Arabnia H. R. (2025) Customer Churn Prediction: A Systematic Review of Recent Advances, Trends, and Challenges in Machine Learning and Deep Learning. Machine Learning and Knowledge Extraction, vol. 7, no 3, art. 105. Available at: https://www.mdpi.com/2504-4990/7/3/105 (accessed April 7, 2026). DOI: https://doi.org/10.3390/make7030105
Khattak A., Mehak Z., Ahmad H., Asghar M. U., Asghar M. Z., Khan A. (2023) Customer churn prediction using composite deep learning technique. Scientific Reports, vol. 13, art. 17294. Available at: https://www.nature.com/articles/s41598-023-44396-w (accessed April 7, 2026). DOI: https://doi.org/10.1038/s41598-023-44396-w
Papa A., Shemet Y., Yarovyi A., Vahovska L. (2022) Development of information technology for analyzing the customer churn of a telecommunication company. Information and Control Systems, vol. 2, no 2(64), pp. 11–15. DOI: https://doi.org/10.15587/2706-5448.2022.255861
Reichheld F. F., Sasser W. E. Jr. (1990) Zero defections: Quality comes to services. Harvard Business Review, vol. 68, no 5, pp. 105–111.
Yevtushenko N. M., Kryvenko Yu. V., Stetsenko D. I. (2024) Tsyfrovi tekhnolohii u strakhuvanni [Digital technologies in insurance]. Mizhnarodnyi naukovyi zhurnal «Hral nauky», no 43, pp. 105–114. DOI: https://doi.org/10.36074/grail-of-science.06.09.2024.011 (in Ukrainian)
Suruzhiu I. S., Leonova S. V., Taranskyi I. P. (2024) Shtuchnyi intelekt u protsesi transformatsii intehrovanykh marketynhovykh komunikatsii innovatsiino aktyvnykh pidpryiemstv [Artificial intelligence in the transformation of integrated marketing communications of innovative enterprises]. Marketynh i tsyfrovi tekhnolohii, vol. 8, no 4, pp. 45–58. DOI: https://doi.org/10.5281/zenodo.14032942 (in Ukrainian)
Copyright (c) 2026 Сергій Кобець, Олена Івасенко
This work is licensed under a Creative Commons Attribution 4.0 International License.
